Open AccessCognitive Experiments and Features for Computing Mental Stress
Author(s) -
Khalid Masood,
Haroon Rashid
Publication year - 2017
Publication title -
lahore garrison university research journal of computer science and information technology
Language(s) - English
Resource type - Journals
eISSN - 2521-0122
pISSN - 2519-7991
DOI - 10.54692/lgurjcsit.2017.010319
Subject(s) - support vector machine , cognition , stress (linguistics) , mental stress , computer science , protocol (science) , artificial intelligence , margin (machine learning) , machine learning , psychology , medicine , psychiatry , linguistics , philosophy , alternative medicine , pathology
In this paper, mental stress is computed through cognitive experiments that induce stress. In a controlled laboratory environment, a group of students are involved in a series of mental challenges. While performing the cognitive tasks, stress is induced on the participants. Deep breathing exercise is performed in the start of experiments and then in between each activity to make the conditions normal and a participant feels relaxed. Various physiological features are recorded during experimental activities. Also, cerebral features are recorded that provide improved classification results. The severity of stress is different on each participant but the purpose of experimental protocol is to separate stressful conditions from relaxed environment. Support Vector machine (SVM) is used to identify relax or normal class from a number of stressed classes. It is shown that cerebral features improve the classification accuracy with a satisfactory margin and designed protocol system is able to compute the severity of induced stress.